mbox series

[0/3] ceph: forward average read/write/metadata latency

Message ID 20220215091657.104079-1-vshankar@redhat.com (mailing list archive)
Headers show
Series ceph: forward average read/write/metadata latency | expand

Message

Venky Shankar Feb. 15, 2022, 9:16 a.m. UTC
Right now, cumulative read/write/metadata latencies are tracked
and are periodically forwarded to the MDS. These meterics are not
particularly useful. A much more useful metric is the average latency
and standard deviation (stdev) which is what this series of patches
aims to do.

The userspace (libcephfs+tool) changes are here::

          https://github.com/ceph/ceph/pull/41397

Note that the cumulative latencies are still forwarded to the MDS but
the tool (cephfs-top) ignores it altogether.

Latency standard deviation is calculated in `cephfs-top` tool.

Venky Shankar (3):
  ceph: track average r/w/m latency
  ceph: include average/stdev r/w/m latency in mds metrics
  ceph: use tracked average r/w/m latencies to display metrics in
    debugfs

 fs/ceph/debugfs.c |  2 +-
 fs/ceph/metric.c  | 44 +++++++++++++++++++++++----------------
 fs/ceph/metric.h  | 52 +++++++++++++++++++++++++++++++++--------------
 3 files changed, 65 insertions(+), 33 deletions(-)

Comments

Xiubo Li Feb. 15, 2022, 11:12 a.m. UTC | #1
On 2/15/22 5:16 PM, Venky Shankar wrote:
> Right now, cumulative read/write/metadata latencies are tracked
> and are periodically forwarded to the MDS. These meterics are not
> particularly useful. A much more useful metric is the average latency
> and standard deviation (stdev) which is what this series of patches
> aims to do.
>
> The userspace (libcephfs+tool) changes are here::
>
>            https://github.com/ceph/ceph/pull/41397
>
> Note that the cumulative latencies are still forwarded to the MDS but
> the tool (cephfs-top) ignores it altogether.
>
> Latency standard deviation is calculated in `cephfs-top` tool.
>
> Venky Shankar (3):
>    ceph: track average r/w/m latency
>    ceph: include average/stdev r/w/m latency in mds metrics
>    ceph: use tracked average r/w/m latencies to display metrics in
>      debugfs
>
>   fs/ceph/debugfs.c |  2 +-
>   fs/ceph/metric.c  | 44 +++++++++++++++++++++++----------------
>   fs/ceph/metric.h  | 52 +++++++++++++++++++++++++++++++++--------------
>   3 files changed, 65 insertions(+), 33 deletions(-)
>
This series looks good to me.

Reviewed-by: Xiubo Li <xiubli@redhat.com>
Jeff Layton Feb. 15, 2022, 11:38 a.m. UTC | #2
On Tue, 2022-02-15 at 14:46 +0530, Venky Shankar wrote:
> Right now, cumulative read/write/metadata latencies are tracked
> and are periodically forwarded to the MDS. These meterics are not
> particularly useful. A much more useful metric is the average latency
> and standard deviation (stdev) which is what this series of patches
> aims to do.
> 
> The userspace (libcephfs+tool) changes are here::
> 
>           https://github.com/ceph/ceph/pull/41397
> 
> Note that the cumulative latencies are still forwarded to the MDS but
> the tool (cephfs-top) ignores it altogether.
> 
> Latency standard deviation is calculated in `cephfs-top` tool.
> 
> Venky Shankar (3):
>   ceph: track average r/w/m latency
>   ceph: include average/stdev r/w/m latency in mds metrics
>   ceph: use tracked average r/w/m latencies to display metrics in
>     debugfs
> 
>  fs/ceph/debugfs.c |  2 +-
>  fs/ceph/metric.c  | 44 +++++++++++++++++++++++----------------
>  fs/ceph/metric.h  | 52 +++++++++++++++++++++++++++++++++--------------
>  3 files changed, 65 insertions(+), 33 deletions(-)
> 

Looks good, Venky. Merged into testing branch. I did make a small change
to the last patch to fix a compiler warning. PTAL and make sure you're
OK with it.

Thanks,
Venky Shankar Feb. 15, 2022, 12:26 p.m. UTC | #3
On Tue, Feb 15, 2022 at 5:08 PM Jeff Layton <jlayton@redhat.com> wrote:
>
> On Tue, 2022-02-15 at 14:46 +0530, Venky Shankar wrote:
> > Right now, cumulative read/write/metadata latencies are tracked
> > and are periodically forwarded to the MDS. These meterics are not
> > particularly useful. A much more useful metric is the average latency
> > and standard deviation (stdev) which is what this series of patches
> > aims to do.
> >
> > The userspace (libcephfs+tool) changes are here::
> >
> >           https://github.com/ceph/ceph/pull/41397
> >
> > Note that the cumulative latencies are still forwarded to the MDS but
> > the tool (cephfs-top) ignores it altogether.
> >
> > Latency standard deviation is calculated in `cephfs-top` tool.
> >
> > Venky Shankar (3):
> >   ceph: track average r/w/m latency
> >   ceph: include average/stdev r/w/m latency in mds metrics
> >   ceph: use tracked average r/w/m latencies to display metrics in
> >     debugfs
> >
> >  fs/ceph/debugfs.c |  2 +-
> >  fs/ceph/metric.c  | 44 +++++++++++++++++++++++----------------
> >  fs/ceph/metric.h  | 52 +++++++++++++++++++++++++++++++++--------------
> >  3 files changed, 65 insertions(+), 33 deletions(-)
> >
>
> Looks good, Venky. Merged into testing branch. I did make a small change
> to the last patch to fix a compiler warning. PTAL and make sure you're
> OK with it.

Looks good. Thanks for the fix.

>
> Thanks,
> --
> Jeff Layton <jlayton@redhat.com>
>